Identifying Latent Similarities among Near-Miss Incident Records Using a Text-Mining Method and a Scenario-Based Approach.
Tetsuo SawaragiKouichi ItoYukio HoriguchiHiroaki NakanishiPublished in: HCI (9) (2009)
Keyphrases
- text mining
- cost function
- high precision
- synthetic data
- support vector machine svm
- similarity measure
- preprocessing
- computational cost
- machine learning
- segmentation method
- detection method
- computationally efficient
- significant improvement
- computational complexity
- databases
- prior knowledge
- high accuracy
- model selection
- high dimensional
- segmentation algorithm
- optimization algorithm
- clustering method
- feature extraction
- decision trees
- genetic algorithm
- web mining